Importing specific columns of a csv file into a 2D Numpy array format

Question

a university assignment has us tasked with constructing code that analyses CSV files and selects certain columns based on their header. Here is the question:

---

Write a function load_metrics(filename) that given filename (a string, always a csv file with same columns as given in the sample metric data file), extract columns in the order as follows:

created_at
tweet_ID
valence_intensity
anger_intensity
fear_intensity
sadness_intensity
joy_intensity
sentiment_category
emotion_category

The extracted data should be stored in the NumPy array format (i.e., produces <class 'numpy.ndarray'>). No other post-processing is needed at this point. The resulting output will now be known as data.

Note: when importing, set the delimiter to be ',' (i.e., a comma) and the quotechar to be '"' (i.e., a double quotation mark).

---

And here is the code ive written so far:

import csv
import numpy as np

def load_metrics(filename):

    """Loads data from csv files"""

    col_list = ["created_at","tweet_ID","valence_intensity",
                "anger_intensity","fear_intensity","sadness_intensity",
                "joy_intensity","sentiment_category",
                "emotion_category"]

    with open(filename, 'r') as csvfile:
        data = np.loadtxt(csvfile, delimeter=",", quotechar='"', usecols=col_list)
    
    return data
    

any improvements I can make? Thank you.

Answer 1

I haven't tried it myself but from the documentation it looks like usecols of numpy.loadtxt expects a sequence of integers

https://numpy.org/doc/stable/reference/generated/numpy.loadtxt.html

You need to find which column numbers are the ones you want out of all the columns